Few-shot Classification on Flowers102 16-shot (test)
94.2Accuracy (Sym 0.125)NA-MVP
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| NA-MVPImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 94.2 | 93.3 | 92 | 90.47 | 85.07 | 76.47 | 91.37 | 78.43 | |
| NLPromptImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 93.87 | 92.57 | 92.73 | 89.9 | 84.77 | 76.8 | 93.4 | 81.1 | |
| CoOpImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 88.93 | 83.5 | 77.93 | 70.1 | 55.6 | 37.17 | 74.7 | 42.6 | |
| GCEImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 88.8 | 88.33 | 86.73 | 84.07 | 78.37 | 70.37 | 86.37 | 69.93 | |
| JoAPRImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 84.9 | 84.7 | 79.75 | 77.13 | 69.65 | 64.2 | 79.57 | 55.47 |